Clojure numerical yak shave: provably efficient deep networks. Best mathy paper I've read lately. https://github.com/lynaghk/lorax /cc @_onionesque
@johnmyleswhite @_onionesque It's a cool algo. The author's implementation conflates test/train process, which took me a while to untangle.
-
-
@lynaghk@_onionesque Early stopping style conflates? Or something worse? -
@johnmyleswhite@_onionesque Train/test are rows in same matrices; cannot just segment because you need to iterate entire matrix to inc dim
End of conversation
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.